Point cloud extraction method, system, equipment and medium based on epipolar image pyramid

By using a method based on the epipolar image pyramid, the optimal disparity is calculated layer by layer to generate a dense point cloud, which solves the problem of lack of global constraints in multi-baseline and multi-view image matching and improves the reliability and efficiency of the matching results.

CN119478235BActive Publication Date: 2025-09-30Chinese People's Liberation Army Cyberspace Force Information Engineering University
2 Cites 0 Cited by

Patent Information

Application Number
CN202411632857.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-11-15
Publication Date
2025-09-30
Estimated Expiration
2044-11-15

AI Technical Summary

Technical Problem

Existing image matching methods lack global constraints in the extraction of dense point clouds from multi-baseline and multi-view images, resulting in gross errors in the matching results. In addition, the SGM image matching method occupies a large amount of memory and has a slow calculation speed.

Method used

A method based on epipolar image pyramid is adopted. The feature points are extracted through the initial SIFT feature matching algorithm. The number of pyramid epipolar image layers is calculated, and the optimal disparity is calculated layer by layer to finally generate a dense point cloud.

Benefits of technology

It achieves highly reliable dense point cloud extraction, reduces gross errors in matching results, reduces memory usage and computing time, and improves matching efficiency.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention provides a point cloud extraction method, system, device, and medium based on an epipolar image pyramid. The method comprises: extracting feature points from a collected stereo epipolar image using an initial SIFT feature matching algorithm, calculating the number of layers of the pyramid epipolar image using the disparity of all extracted feature points; calculating the optimal disparity of each layer from the top image to the middle image based on the bottom-level disparity of the bottom-level feature points in the bottom-level image of the pyramid epipolar image, iteratively calculating the optimal disparity of the bottom-level feature points based on the optimal disparity of each layer; matching the current collected stereo epipolar image using the optimal disparity of the bottom-level feature points to obtain feature points with the same name, and generating a dense point cloud based on the feature points with the same name. The method of the present invention takes into account the overall information of the image and is a matching method with global constraints, capable of obtaining a highly reliable dense point cloud.
Need to check novelty before this filing date? Find Prior Art